Gemini 3.1 Pro Preview High: LiveBench scores and how to run it with BrainOutput
Gemini 3.1 Pro Preview High is a model from Google. In the LiveBench release of 2026-06-25 (fetched 2026-10-03) it ranks #26 of 66, with a global score of 76.95.
What is Gemini 3.1 Pro Preview High?
Gemini 3.1 Pro Preview High is listed on LiveBench, a public, continuously refreshed benchmark that scores models on seven categories: reasoning, coding, agentic coding, mathematics, data analysis, language and instruction following. Its strongest category in this snapshot: instruction following (#3 of 66).
Its weights are not published: it is available only through its vendor's service. With BrainOutput it stays a cloud option for non-sensitive work, when your policy allows it.
The model at a glance
| Vendor | |
|---|---|
| Weights | Proprietary (weights not published) |
| Licence | proprietary (API only) |
| Release date (per LiveBench) | — |
| LiveBench identifier | gemini-3.1-pro-preview-high |
LiveBench scores, dated
Score and rank out of 66 models, per category, in the LiveBench release of 2026-06-25 fetched 2026-10-03.
LiveBench, release 2026-06-25 · livebench.ai · 66 models · fetched 2026-10-03
Overall rank in this LiveBench snapshot.
Global score: the mean of the seven categories.
Gap to #1, Claude Fable 5.1 Max Effort (83.41).
| Category | Score | Rank |
|---|---|---|
| Reasoning | 84.00 | #40 / 66 |
| Coding | 76.45 | #43 / 66 |
| Agentic coding | 44.14 | #53 / 66 |
| Mathematics | 91.05 | #29 / 66 |
| Data analysis | 78.54 | #23 / 66 |
| Language | 85.38 | #14 / 66 |
| Instruction following | 79.10 | #3 / 66 |
| Global (mean of the 7 categories) | 76.95 | #26 / 66 |
Method note: the global score is the mean of the seven category averages, computed by us from LiveBench's published file; it may differ by a few decimals from the site's own display. Equal scores keep LiveBench's file order. Benchmarks measure benchmark tasks — during an AI diagnostic we re-run the comparison on your own workloads.
Running Gemini 3.1 Pro Preview High with BrainOutput
BrainOutput routes each task to a model by confidentiality: sensitive work stays on a local or private model, and everything else goes to a cloud model only when your policy allows it.
Locally, on your hardware
Not possible
Not possible: the weights are not published. Sensitive work goes to an open-weight model running locally instead.
In a private cloud
Not possible
Not possible as private inference: only the vendor serves this model.
With your own key (BYOK)
Available
Plug in your own Google API key: BrainOutput routes non-sensitive work to it when your policy allows; sensitive work never leaves your perimeter.
Frequently asked questions
Is Gemini 3.1 Pro Preview High an open-weight model?
No: its weights are not published. The highest-ranked open-weight model in this snapshot is DeepSeek V4.1 Flash Max Effort (#9, 81.11).
Can we use Gemini 3.1 Pro Preview High with BrainOutput?
Yes, through your own Google API key (BYOK), for non-sensitive work and only when your policy allows it. Sensitive work stays on an open-weight model running locally or in a private cloud.
Where do these scores come from?
From LiveBench (livebench.ai), release 2026-06-25, file fetched 2026-10-03. The global score is the mean of the seven categories, computed by us. The ranking changes monthly; we re-verify at diagnostic time.
Find out whether Gemini 3.1 Pro Preview High fits your work
An AI diagnostic maps your processes, classes what is sensitive, and benchmarks current models against your real tasks — dated numbers, yours this time.
Sources
- LiveBench — leaderboard, release 2026-06-25 — accessed 2026-10-03